Saturday, 25 April 2020
A report from an American health institution predicting that Belgium will suffer zero deaths from the coronavirus (Covid-19) after May 19 has been played down by Belgium’s own experts on the epidemic.
The prediction was made by the Institute for Health Metrics and Evaluation (IHME) in Seattle, Washington state, using a model created by and named after its director, Chris Murray. The model was developed to forecast the impact of the global pandemic on health-care systems in the US.
The institute also applied it to the case of Belgium, using the official statistics produced daily by the government’s Covid-19 crisis centre and the health institute Sciensano.
The Belgium report was created on April 21, when it projected the number of deaths would be 170. In fact the number – reported on April 22 – was 259.
For the days between then and now, the model predicts the number of deaths (real figures in brackets) as: 241 (230), 223 (248) and 203 (183). The actual figures have been adjusted to take account of a delay in reporting from some Flemish care homes which led to 58 deaths being reported 24 hours late.
So the model seems to be fairly accurate. But then it predicts a precipitous drop in the number of fatalities from April 22, falling to 53 on May 4, the first day of the new phase. It then falls to 21 on May 8, 12 on May 10 and 5 on May 13, losing one a day to reach one a day on May 17 and zero by May 19.
“I hope it is realistic, but it remains a scenario and a model,” said Professor Steven Van Gucht of the crisis centre. “We have to wait and see what the facts will say. It is true that most people who get infected experience a mild infection. A small minority die. When there is still little of the virus and little infection in circulation, the number of deaths will become rarer. No deaths at all in the coming weeks or months? I’d be surprised.”
Sudinfo asked virologist Professor Marc Van Ranst if the Chris Murray model results were at all realistic.
“It’s hard to say,” he replied. “For Belgium, the model sounds good for this phase of the epidemic, but the quality of the data remains too low to really compare with other countries, since the mortality rate here is calculated differently.”
And he pointed out that the view shared among epidemiologist that a zero death count does not by any means signal the end of the matter. The common view among experts is that the virus will not disappear, but its ability to infect large numbers will be reduced until the right circumstances come along for a revival.
“There will be another difficult period in the autumn and next winter, as new infections begin to increase thanks to the weather conditions,” said Prof Van Gucht.
The Brussels Times